Skip to content

The code finds the edges on the Musical note images first by performing convolution with a sobel filter. Then it tries to find the coordinates of the notes occur using template matching. It finds the location of staff lines using Hough transform. With this data, it estimates the correct note.

Notifications You must be signed in to change notification settings

manikandan5/Optical_Music_Recognition

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

51 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Optical Music Recognition

This code was written for the course Computer Vision (CSCI-B 657) at Indiana University handled by Professor David Crandall. Skeleton code was provided by the Professor to get us started with the assignment.

What does the program do?

  • The program tries to find the musical notes from the music sheet.

How does it find it?

  • Initially we try to remove the noise in the image by convoluting the image with a Gaussian filter.
  • Then we find the edges in the image by convoluting with a Sobel filter.
  • Then the staff lines are found using Hough Transformation.
  • Once the staff lines are found, we find on which staff line the nodes are present using template matching.
  • With this data, the program tries to estimate the correct note.

Detailed explanation about how the code works and the reason why we chose this implementation could be found here.

How to run the program?

This command compiles the program:

  • make

To run the program:

  • ./omr image.png

About

The code finds the edges on the Musical note images first by performing convolution with a sobel filter. Then it tries to find the coordinates of the notes occur using template matching. It finds the location of staff lines using Hough transform. With this data, it estimates the correct note.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •